Skip to content

Start free with 1,000 credits - no credit card requiredSee plans

VelATL
Log inStart Free

Migration guide

From OpenRouter in one line.

Same OpenAI shape, shorter model names, INR billing. For whether you should switch at all, read the honest comparison first - this page is the how.

1. Swap the baseURL

Before / after - OpenAI SDK

# Before (OpenRouter)
client = OpenAI(
    base_url="https://openrouter.ai/api/v1",
    api_key="sk-or-...",
)
model = "openai/gpt-oss-120b"  # long provider/model strings

# After (VelATL)
client = OpenAI(
    base_url="https://velatl.ai/api/v1",
    api_key="velatl_...",  # Settings -> API keys, capped at 5,000 cr/mo by default
)
model = "balanced"  # or "fast" - short slugs, same everywhere

OpenCode, Cline and Continue need only the provider URL and key changed - see the 1-minute OpenCode setup. Claude Code users: swap ANTHROPIC_BASE_URL the same way (guide).

2. Map the model names

  • OpenRouter addresses every model as provider/model; VelATL exposes curated slugs (fast, balanced) with published per-1k credit rates.
  • GET /api/v1/models returns your plan's models with rates - use it to validate names at startup instead of hardcoding.
  • Trade-off, stated plainly: you lose the 200-model zoo and per-request provider choice. You keep streaming, tools and JSON mode (text + tools only - limits).

3. Move the billing

  • OpenRouter bills USD pay-as-you-go with minimum purchases; VelATL bills INR plans plus top-ups via Razorpay, and credits never expire.
  • Set one key per project with its own monthly cap in Settings - the closest equivalent to OpenRouter's per-key limits, counting settled + in-flight spend.
  • Every VelATL response carries X-Request-Id; every charge lands in one ledger with automatic refunds on failure - reconcile against it, not the dashboard.
Start free with 1,000 creditsCompare first